perf: Allow ftracing perf code
Subsystems (2)
PATCH 1/1 perf: Allow ftracing perf code
From: Andi Kleen <[email protected]>
According to Stephen there is no reason left not to allow function trace in perf code. Don't disable the function tracer instrumentation for the perf files. Changes the generic and the x86 files.
I've been using variants of this patch for a long time and haven't seen any problems. It's really useful to figure out why a complex perf_event_open() syscall fails.
--- arch/x86/kernel/cpu/Makefile | 1 - kernel/events/Makefile | 4 ---- 2 files changed, 5 deletions(-)
arch/x86/kernel/cpu/Makefile +0 -1
@@ -5,7 +5,6 @@ # Don't trace early stages of a secondary CPU boot ifdef CONFIG_FUNCTION_TRACER CFLAGS_REMOVE_common.o = -pg -CFLAGS_REMOVE_perf_event.o = -pg endif # Make sure load_percpu_segment has no stackprotector
kernel/events/Makefile +0 -4
@@ -1,7 +1,3 @@ -ifdef CONFIG_FUNCTION_TRACER -CFLAGS_REMOVE_core.o = -pg -endif - obj-y := core.o ring_buffer.o callchain.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
Signature · 1 line
1.8.3.1